AI Provider Support
RoostGPT supports multiple AI providers. Provider availability varies by test type.
Unit, API, Functional & Integration Tests
OpenAI
| Model | Context Window | Best For | Key Features |
|---|---|---|---|
| GPT-5 | 272K input / 128K output | Complex test scenarios, enterprise | Expert-level intelligence, built-in reasoning |
| GPT-4.5 | Large | Standard test generation | Balanced performance |
| GPT-4o | Large | Fast test generation | Multimodal, real-time processing |
| GPT-4o mini | Standard | High-volume generation | Cost-efficient, fast |
| GPT-4 Turbo | 128K | Production environments | Proven reliability |
| o3 | Extended | Complex reasoning tasks | Advanced chain-of-thought |
| o4-mini | Standard | Cost-efficient reasoning | Faster reasoning model |
Recommended: gpt-4o, gpt-5
Anthropic Claude
| Model | Context Window | Best For | Key Features |
|---|---|---|---|
| Claude Opus 4.1 | 200K in / 32K out | Complex coding, unit tests | Best coding model, 74.5% SWE-bench |
| Claude Sonnet 4.5 | 64K out | Coding, Finance, CyberSecurity | Long-running tasks, 77.2% SWE-bench |
| Claude Sonnet 4 | 200K–1M / 64K out | General test generation | Superior coding and reasoning |
| Claude Sonnet 3.7 | 200K / 64K out | Hybrid reasoning | Fast and extended thinking modes |
| Claude Sonnet 3.5 | 200K / 8K out | Legacy support | Strong coding |
| Claude Haiku 3.5 | 200K / 8K out | Quick, simple tests | Fast, cost-effective |
Recommended: claude-sonnet-4-5
Azure OpenAI
Provides the same OpenAI models with enterprise features (VNET, private endpoints, managed identity, SLA guarantees).
| Model | Context Window | Additional Features |
|---|---|---|
| GPT-5 Series | 272K / 128K | VNET integration, private endpoints |
| GPT-4.5 | Large | Managed identity, regional deployment |
| GPT-4o Series | Large | SLA guarantees, Azure Monitor |
| GPT-4 Turbo | 128K | Standard and provisioned deployments |
| o3, o4-mini | Extended/Standard | Azure AD authentication |
Recommended: gpt-4o, gpt-5
Google Gemini
| Model | Context Window | Best For | Key Features |
|---|---|---|---|
| Gemini 2.5 Pro | 1024K in / 64K out | Complex API testing, large codebases | Deep Think mode, long context |
AWS Bedrock
AWS Bedrock provides access to foundation models in a fully managed service.
| Provider | Models | Context Window | Best For |
|---|---|---|---|
| Anthropic | Claude Opus 4.1, Sonnet 4, Sonnet 3.7 | 4K output | Code generation, unit tests |
See AWS Bedrock Configuration for IAM setup and pricing.
UI Tests
UI tests require vision-capable models. Only the following providers are supported.
OpenAI (UI)
| Model | Best For | Capabilities |
|---|---|---|
| GPT-5 | Complex UI workflows, enterprise apps | Expert intelligence, superior visual perception |
| GPT-4.5 | Standard UI testing | Vision-enabled, balanced performance |
| GPT-4o | Multimodal test generation | Real-time visual analysis, fast response |
| GPT-4o mini | High-volume generation | Cost-efficient, fast, good vision |
| o3 | Complex test logic, deep reasoning | Chain-of-thought with vision |
| o4-mini | Cost-efficient reasoning | Faster reasoning with vision |
Recommended: gpt-4o, gpt-5
Google Gemini (UI)
| Model | Context Window | Best For | Capabilities |
|---|---|---|---|
| Gemini 2.5 Pro | 1024K in / 64K out | Complex UI testing, large apps | State-of-the-art thinking, long context |
| Gemini 2.5 Flash | 1024K in / 64K out | Fast, cost-effective | Best price-performance |
| Gemini 2.5 Flash-Lite | 1024K in / 64K out | High-volume, low-latency | Most cost-effective |
| Gemini 2.0 Flash | 1024K in / 8K out | Modern apps | Native tool use, improved speed |
| Gemini 2.0 Flash-Lite | 1024K in / 8K out | Budget projects | Cost-efficient, low latency |
Recommended: gemini-2.5-pro, gemini-2.5-flash
Azure OpenAI (UI)
Same vision-capable models as OpenAI with Azure enterprise features.
| Model | Context Window | Additional Features |
|---|---|---|
| GPT-5 Series | Extended | Azure security, compliance, VNET |
| GPT-4.5 | Large | Private endpoints, managed identity |
| GPT-4.1 Series | Large | Regional deployment, data residency |
| GPT-4o Series | Large | SLA guarantees, Azure Monitor |
| o3, o4-mini | Extended/Standard | Azure AD authentication |
Recommended: gpt-4o, gpt-5